So when making the 100% TAS, I actually analyzed this route to get to the green frog and it was slower than the route that was settled on. (I checked my notes from August 13, 2013 to verify)

Using the first frame of the fade to dark after menu select is basically in line with what I did. The first frame of the fade in your "53.99" run would be the 1614th Frame (13.03s). So a time of 54s would be accurate.

We don't currently have a co-op any% category but one could be made.

I wanted to post a thread to discuss this. The exact start and stop time for runs isn't typically critical to get exact frame accuracy on, but since raids_shine submitted the following Any% with Warp run, I think it is important to state the exact start & finish time.

The start time is from menu select:

This frame is from exactly the 13.00 second mark [Frame #312] but based on audio I'm not sure if the exact frame would be this or the next frame.

The time stops when the text "C L E A R" appears:

This frame is at 1:07.21 [Frame #1613].

Thus, this run is 54.21s long.

I wanted to get other opinions since this run seems basically optimal so I'm not sure if 53s is possible. Many years ago when selecting the end of the run, the initial idea we had was when the play loses control which would be a few frames prior to the CLEAR text, but we chose the CLEAR text as that is a more consist method. If the start & finish times are adjusted a few frames, menu select to lose control, this run could potentially be frames faster. At present the run would need to be 6 frames faster to be 53.xx long.

I went ahead and added Original Mode as a Misc category. Let me know if I should adjust anything.

Adrian's formula only applies to Intelligent Qube. I've attempted to determine I.Q Final's formula as it is different. The formula I posted above is more accurate but there's a few known shortcomings. Data

• The same machine code addition error from I.Q is present in I.Q Final. So barrier cases can round inconsistently. Adrian explains the machine code addition error on his site.

• There appears to be an additional factor that is only present when you clear multiple stages. Thus I.Q Final's equation is not isolated to the present stage's performance, but rather can be influenced based on your overall performance. This I have not yet been able to account for.

The whole purpose of this math is to:

1. Know in the later stages, specifically the final stage, how well you need to preform to get 500 I.Q. If I.Q can be accurately solved for, you will be able to know in the Final stage the score/perfects required to get 500 I.Q as well as have an average performance to use as your baseline.

2. Be able to develop accurate statistically trends per stage so you can have more accurate estimates of how well you need to perform early game. For the later stages, Any% clears will still have score gains naturally from them, so using statistical analysis on an aggregate of runs will provide us with a average range of I.Q points that will be achieved. Thus a more accurate assessment of early game performance, as well as measuring the current run's performance, will be achievable.

Obviously not necessary for just doing runs, but something I wanted to solve before going for a more optimized run, as just hoping you get 500 I.Q is rather frustrating.

Absolutely!

This is a category I've wanted to run, but I've never been able to crack the equation for I.Q. I did a bunch of math on this but was never able to get an accurate equation.

• • Math - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

The closest I've gotten is the following equation:

I.Q = D * S * (A + B * (P + E)) / 10000

D = Difficulty Factor { Normal = 1, Ultra Hard = 1.2}

S = Score

A = Stage Parameter {stage 1 = 5, stage 2 = 4.8, stage 3 = 2.5, stage 4 = 2.35, stage 5 = 2.2, stage 6 = 2.05, stage 7 = 1.85, stage 8 = 1.65, Final = 1.55}

B = Completion Parameter {stage 1 = 3, stage 2 = 2.88, stage 3 = 1.5, stage 4 = 1.41, stage 5 = 1.32, stage 6 = 1.84, stage 7 = 1.11, stage 8 = 1.49, Final = 0.7}

P = # of Perfect Clears on a stage

E = # of Excellent Clears on a stage

Using this, I am able to calculate I.Q on any individual stage, but when you use this formula across an entire run, it is no longer accurate and I've never managed to account for this. So the formula is incomplete.

• • End of Math - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

I am curious if you have a solution for this, or what method you are using to predict your I.Q before you beat the game to ensure you have 500 I.Q?

SRC is not structured in a way to comfortably host these records in the way the IL community desires. So those are tracked independently of this website.

The IL records are tracked here:

https://docs.google.com/spreadsheets/d/1y5SN4rHPr4w16SScHoYo7nN63LMZA_f-bQ6RZv0JdI8/edit#gid=930287183

I have added this as a new category. I called it "Any% w/ Warps". Is that an acceptable name?

raids_shine is correct. PS1 emulators are attempting to replicate the PS1 hardware accurately. This is a PS2 feature, so it is not possible on emulator.

Yes. For most PS1 games, enabling Fast disc speed is standard. The only exceptions generally are games where it can crash the game, which is not a problem for Intelligent Qube or I.Q Final.

I'm confused on how this works. Is this the practice mode taking you to the Final stage in order to allow you to practice the block configuration selected?

I don't know that I really understand the purpose of the category you're requesting. It would be a combined score for P1 & P2 to get 2/5,000,000 points. It feels like a fairly contrived objective, but that isn't inherently disqualifying. A similar objective to this is in Puyo Puyo Champions, it has several score speedrun leaderboards in the single player practice area so speedrunning to X amount of points is a reasonable objective. Of course, that's single player, the co-op aspect of this request makes it odd.

I guess to me, I'd be more open minded about starting with a leaderboard for single player large score objectives if that's something that can be done within the game. You all are beginner players so I would recommend looking up material on how to build chains in Puyo Puyo, as this game has strong similarities in strategy. The blobs stretching after pops is the primary difference and is notable once you start building for longer chains (5+) but I didn't see any lengthy chains from the part of your video I did skim.

Just my thoughts.

That sounds more like a score style challenge run than a time based speedrun. You're more than welcome to attempt this if you're interested, but I don't see the value in having a speedrun leaderboard for it.

The IL records are tracked here: https://docs.google.com/spreadsheets/d/1y5SN4rHPr4w16SScHoYo7nN63LMZA_f-bQ6RZv0JdI8/edit#gid=930287183

That's not the end of the game.

Also, this game was released 11 December 2003 fyi, not 2004

Can you add console options for the leaderboard? PS2, PS3

Also would tracking in game time be worth while?

Console & region options don't split the leaderboard, just adds more information. Also emulation rules are pretty standard for playstation speedrunning.